2013 pre-ChatGPT text flagged by AI detector Pangram, false positive exposed
TuhinChakr · x · 2026-09-23
A French literary text written in 2013 — a decade before ChatGPT — was tested against the AI-detection tool Pangram and judged not human-written. The poster clarifies it wasn't the author's style that fooled Pangram, and the text's pre-ChatGPT provenance is verifiable, adding to mounting evidence that AI detectors produce false positives on human authors.
Related event: AI Detector Pangram Flags Decade-Old Human Text as AI(3 posts)→
More from Fun
- Applicant uses Claude to write their resume for an Anthropic job — rejection would mean fumbling its own model — theteknosaur · 2026-11-11
- Fable 5.2 wows with game-generation: detailed xianxia scene built from tools alone — op7418 · 2026-11-11
- Vibes with agents feel like being a Green Lantern wielding the ring — rachittshah · 2026-11-11
- A 3D ray tracing engine written entirely in SQL — fofrAI · 2026-11-11
- Splash One: flight combat sandbox built with Opus 5.5 in PlayCanvas — willeastcott · 2026-11-11
- PlayCanvas engine repo hits 10,000th issue/PR — and it's a Claude-authored merge — willeastcott · 2026-11-11